Posted by Richard Solomon on December 20, 2013
I guess Heraclitus’ real quote is more like “Nothing endures but change” but I’m going to claim poetic license here and stick with “The only constant is change” – because I like the apparent contradiction better. Two events this week drove home the aptness of this quote to me. First we had the announcement that my previous employer LSI was being purchased by Avago. I’m not much of a business person, so I’m not sure quite how to interpret “drive LSI’s operating margins toward Avago’s current levels” but I do know how to interpret a stock price jump from around $8/share to around $11/share! (I believe the technical expression there is “Woo hoo!”) Second, we have a raft of internet news sources reporting that Calxeda is restructuring (or shutting down, depending on whose account you read). I don’t read “restructuring” (as Calxeda is quoted) quite the same as “shutting down”, but hey, we already established that I’m not a business guy. I guess I can’t complain too much about that use of poetic license though since I just exercised it at the top of this posting.
The LSI-Avago deal may not have a huge impact on our industry, beyond the employees and stockholders of course, but it hits close to home for me personally. However, it would be hard to overlook the impact that Calxeda has had on the course of servers in the last five or so years. While there are several companies developing in the ARM server space, Calxeda was one of the first and one of the few that appeared to already be successful in getting silicon into real products. Much ink (and electrons) will undoubtedly be spilled in the weeks ahead arguing whether this signals the downfall of yet-another-x86-competitor in the server space or not. (Does anyone else remember the Advanced RISC Computing standard?) I would argue that more importantly it shows the ripples of change are spreading throughout the server industry.
I’m not sure the origin of the statement that “A pioneer is someone lying face down in a field with arrows stuck in his back” but it seems a bit apt here. Back when Calxeda was founded you would have been hard-pressed to find anyone worried about server power consumption. “Servers care about three things” we’d say “performance, performance, and performance!” Today in contrast one finds data centers located near icy bays to ensure cooling! Suddenly power consumption in servers became almost as important as performance – well, maybe even MORE important than performance. ARM moved from what “everybody knew” to be a mobile CPU vendor, to a contender in the 64-bit multi-core “serious server” CPU space. Even mighty Intel is building server-class CPUs whose focus is at least as much on low-power as it is high-performance. I think it’s clear that pressure from folks like Calxeda helped bring on that particular change! Whatever your opinion on exactly how we’ve come to this point of power-efficiency focus, now more than ever “the only constant is change.”
Even the PCI Express interface, which is almost universally used in servers is undergoing change – as Scott and I have been talking about M-PCIe here for months now. So whatever CPU your next great server-on-a-chip (or any) design uses, and whether you support PCIe or M-PCIe (or both), you can count on us for the IP you need to make that project a success. Ok, so some things never change 🙂 I suppose this is as good a time as any to mention that the Synopsys DesignWare IP for PCI Express can also provide a native interface to various AMBA interfaces found on those ARM server CPUs to connect to either PCI Express or M-PCIe.
Please leave me a comment below with your opinion on ARM servers, low-power servers, or even just applicable quotes from history. As always, thank you for reading, and if you haven’t already done so, please subscribe to ExpressYourself .
I’ve been involved in the development of PCI chips dating back to the NCR 53C810 and pre-1.0 versions of the PCI spec so have definitely lived the evolution of PCI Express and PCI since the very beginning! Over the years I have worked on variations of PCI, eventually moving on to architecting and leading the development of the PCI Express and PCI-X interface cores used in LSI’s line of storage RAID controller chips. For the last ten plus years I've also had the honor of serving on the PCI-SIG Board of Directors and holding a variety of officer positions. I am still involved in PCI-SIG workgroups and I present at many PCI-SIG events around the world. Back before the dawn of PCI, I received my B.S.E.E. from Rice University, and hold over two dozen U.S. Patents, many of which relate to PCI technology.